Effect of silver doping on the band gap tuning of tungsten oxide thin films for optoelectronic applications

Abstract

In the cutting-edge world, semiconductor metal oxides usually tend to have a high optical band gap (>3.0 eV), significantly acceptable for potential optoelectronic applications. The present study discusses the synthesize of pristine tungsten trioxide (WO₃) and Silver (Ag) doped WO₃ (Ag: WO₃) thin films onto a glass substrate at 450 °C, with varying concentrations of Ag doping (2, 4, 6, 8 and 10 at.%) using a simple Spray Pyrolysis Technique. Field emission scanning electron microscopy (FESEM) analysis showed the presence of particles in the WO₃ and Ag: WO₃ materials. The X-ray diffraction (XRD) pattern confirmed that the samples' hexagonal structure remained intact. In addition, Rietveld refinement was used for the samples to study the crystal structure meticulously. Because of the surface plasmon resonance effect, the samples' distinguishing characteristics were visible in their optical nature. For pristine WO₃, the experimental band gap was determined to be 3.20 eV, and for varying doping concentrations, it was found to be 3.15 eV–2.90 eV, respectively. Furthermore, the fracture has remained imperceptible at elevated concentrations, resulting in a substantial influence on the optical characteristics of 10% Ag: WO₃ thin films. The estimated redox potential for 2% Ag: WO₃ shows a considerable influence of the band edge potential of the Conduction Band (CB) and Valance Band (VB). The activation energy was determined using temperature-dependent electrical resistivity and exhibited an ohmic nature. The synthesized material exhibited a negative temperature coefficient (NTC) effect at higher concentrations of doping, suggesting its potential applicability as a thermistor. A comprehensive analysis of this present study indicates that Ag can be a viable candidate for doping on WO₃ thin films for use in optoelectronic devices.

Graphical abstract

Highlights

• •High lights for review:
• •we reported the influence of Ag doping on the surface morphology, structural, optical and electrical properties of tungsten trioxide (WO₃) thin films.
• •The measured band gap for pristine WO₃ is found 3.20 eV and then reduced for different concentrations of Ag doping.
• •Narrowing of the band gap and edge of the VB and CB lead to an effective catalyst that could be useful in photo-degradation and a wide range of sensing properties.
• •An effective heat sensing ability is facilitated by negative temperature coefficient in Ag: WO₃ thin films.

1. Introduction

The advancement of modern technology is dependent on technical innovation of thin film industry and has a significant impact from the twentieth century for extraordinary performance in solid-state electronics. Thin films' characteristics differ significantly from those of their bulk counterparts. The improvement of optoelectronic devices is the primary goal of the current study, which focuses on creating metal oxide thin films. Tungsten trioxide (WO₃) has numerous applications in microelectronics, selective catalysis, environmental engineering, solar cells, gas sensors, photoconductivity, photo-electrochemical cells for solar energy conversion, etc. [[1], [2], [3], [4], [5], [6], [7], [8], [9]]. A comprehensive understanding of the structural property and constancy of WO₃ thin films is indispensable for practical device applications. Low-dimensional materials' properties and applications are predominantly determined by their chemical composition, crystal structure, surface morphology and phase stability. The phase transition of WO₃ thin films and nanostructures is induced by heating or annealing is essential for specific purposes [10]. Understanding the various phase transitions of WO₃ can be attained through an experimental and theoretical approach.

WO₃ is an n-type semiconductor metal oxide with a band gap range of approximately 2.7–3.0 eV and visible light irradiation. WO₃ has a hexagonal crystal structure with lattice constants of a = 7.298 $\AA$, and c = 3.899 $\AA$ and several studies found a monoclinic structure as well [11]. At ambient temperature, the band gap of WO₃ is 2.60 eV. Point defects resulting from oxygen vacancies in sub-stoichiometric compounds WO_1-x ($0<\mathrm{x}<1$) significantly impact its electronic properties. Because of its large band gap and remarkable electron-transport capability, WO₃ is a promising material for usage in a wide variety of optoelectronic devices [12]. In smart windows, large-area information displays, energy-saving, and other applications, films demonstrate remarkable electrochemical stability and are very transparent in the visible spectrum. The analysis of the UV spectrum revealed that, depending on the temperature of the substrate, WO₃ has a visible-light transmission of around 70% on average. There is a significant visual transmission at high deposition temperatures [13].

WO_3, is a non-toxic transition metal oxide; in addition, the elements that constitute WO₃ are cheap and plentiful. Due to interfacial energy, two-dimensional growth is preferred between WO₃ and other oxide substrates, leading to high-quality films at reduced temperatures. Sol-gel, electro-spinning, hydrothermal, thermal evaporation, microwave irradiation, spray pyrolysis, sputtering, and electron beam evaporation techniques have been used to prepare WO₃ thin films [[14], [15], [16]]. There are benefits and drawbacks to every approach. One of the easiest methods is Spray Pyrolysis Technique (SPT), a non-vacuum system to deposit thin films over large areas with perfectly doped of different transition materials. Diverse transparent metal oxides thin films i.e. SnO₂, TiO₂, Fe₂O_3, ZnO have been studied using this technique with different doped materials [[17], [18], [19], [20]]. Moreover, to study or develop new properties, pristine WO₃ has been deposited with an alloy of different 3d transition metals [21,22]. Doped WO₃ thin films reveal transformed structural, electrical and optical characteristics.

Silver (Ag) is an excellent conductor of both heat and electricity; and has a white color, a soft luster, and a high degree of ductility and malleability. As a result, Ag has been extensively reported to increase the photo catalyst's response to visible light. The ionic radius of Ag (172 p.m.) is lower than that of W (210 p.m.), suggesting the development of small grains. The effective surface area of gas-sensing components increases as grain size decreases [23]. The energy band gap of silver doped nanoparticles shrank with growing Ag content, as shown by absorption spectra. The absorbance band gap in Ag-doped thin films moves to the red as the amount of Ag rises. In pristine WO₃, Ag dopant significantly impacts structural, optical, morphological, and electrical properties. The formation of grain size can change due to the incorporation of Ag thin films and finely tune the band gap of variety of oxide materials i.e. WO₃, ZnO, etc., [24,25]. However, adjusting the band edge potential can enhance the water-splitting process. Insight into the properties of WO₃ composites could lead to new uses for the material with Ag dopant. The present investigation is primarily concerned with the area of optoelectronic applications. This research investigated the synthesis of thin films of WO₃ and Ag: WO₃ through SPT, with the purpose of using them in optoelectronic devices. To comprehend the effect of Ag content on WO₃ thin films, structural, optical, and electrical characterizations were performed on all potential samples and observed the band gap tuning, edge potential, and negative temperature coefficient.

2. Experimental analysis

2.1. Reagents and film preparation

As a precursor of W and Ag sources, reagent grade tungstic acid [H₂WO₄] (Sigma Aldrich) and silver nitrate [AgNO₃] (Merck, Germany) have been used. The SPT has been used to deposit films of pristine WO₃ and Ag: WO₃. For the synthesis of WO₃ thin films, 0.1 M solution of pristine WO₃ was produced by mixing distilled water and H₂WO₄ in a predetermined ratio. A schematic diagram of the experimental set up and the deposition parameters are shown in Fig. 1 and Table 1.

Fig. 1.

Schematic illustration of a low-cost experimental spray pyrolysis technique set up (Homemade).

Table 1.

Optimize deposition conditions to synthesize pristine WO₃ and Ag-doped WO₃ thin films by Spray Pyrolysis Technique.

Deposition parameters	Optimum value
Temperature	450 ± 5 °C
Concentration	0.1 M
Doping concentration	2.0–10 at. %
Rate of spray	5 mL/min
Air pressure	0.6 bar
Distance between nozzle and substance	25 cm

Then, to dissolve H₂WO₄, a few drops of ammonia solutions are added to the solution. 2.4985 g of H₂WO₄ are combined with 100 mL of distilled water to prepare 0.1 M solution. Contrarily, 2, 4, 6, 8, and 10 at.% Ag: WO₃ solutions were made by dissolving the calculated amounts of H₂WO₄ and AgNO₃. After 1.5 h of stirring, filtration process has been done by filter paper for separation of suspended solid matter from solution.

2.1.1. Possible reaction

Ammonium Para-tungstate was created when H₂WO₄ was dissolved in purified water and ammonia solution. Then, WO₃ produced through the pyrolysis of ammonium para-tungstate.

Moreover, calculated amount of AgNO₃ added for preparing solution of different doped concentration. AgNO₃, H₂WO₄ (precursor), ammonia solution, and water (reactant) combined reaction that may occur on a heated substrate and lead to thin coatings. To produce WO₃ thin film on a heated substrate, the following reaction can take place between H₂WO₄ (precursor), ammonia solution, and water (reactant) [26,27]. As deposited thin films are shown in Fig. 2.

$$H_{2}W{O}_4+{NH}_{4}OH+H_{2}O\to {\left(N{H}_4\right)}_{10}\left(H_2{W}_{12}{O}_{42}\right).4H_{2}O$$ (1)

Then pyrolyzed it,

$${\left(N{H}_4\right)}_{10}\left(H_2{W}_{12}{O}_{42}\right).4H_{2}O\stackrel{450\pm 5°C}{\to }W{O}_3+N{H}_3\uparrow +H_{2}O\uparrow$$ (2)

Fig. 2.

Snapshot of deposited thin films.

Chemical reactions may take place during the formation of doped samples are described below. Distilled water reacts with silver nitrate to generate aqueous silver nitrate.

$$AgN{O}_3\left(s\right)+H_{2}O\left(l\right)\to AgN{O}_3\left(aq\right)$$ (3)

AgNO₃ decomposes when heated,

$$AgN{O}_3\left(aq\right)\to Ag+O_2\uparrow +N{O}_2\uparrow$$ (4)

The probable chemical reactions of Ag: WO₃ are given below,

$$\text{AgN}{O}_3+{\left(N{H}_4\right)}_{10}\left(H_2{W}_{12}{O}_{42}\right).4H_{2}O\stackrel{450\pm 5°C}{\to }\text{Ag}:W{O}_{3+}N{H}_3\uparrow +H_{2}O\uparrow$$ (5)

2.2. Characterizations and techniques

The overall surface morphologies and elemental compositions of the as-deposited films were investigated using a FESEM (Model: JEOL JSM-7600F) and an accompanying EDX detector. The UV–visible spectrophotometer (SHIMADZU UV-1601) was employed to record the UV–visible NIR transmission bands at room temperature. In addition, the Fizeau fringes method is used to observe the thickness of films. Likewise, the Four Point Probe method has been utilized to evaluate electrical properties. The Rietveld refinement was carried out with the Fullprof Suit software. All the graphs were plotted using the Origin software. The ImageJ software was utilized to ascertain the magnitude of the crack's thickness on the film's surface.

3. Result and discussions

3.1. Structural analysis

The XRD technique was employed to perform a structural investigation of Ag: WO₃ samples, where 0, 2, 4, 6, 8 and 10% Ag concentrations were examined. Fig. 3 displays the outcomes of the Rietveld refinement conducted on samples of Ag: WO₃. The pristine WO₃ exhibits diffraction peaks at 2 $\theta$ values of 13.98°, 23.32°, 24.30°, 27.28°, 28.12°, 33.58°, 36.86°, 49.78°, and 55.67° which are indexed to the corresponding diffractions from (100), (001), (110), (101), (200), (111), (201), (220) and (221) planes, respectively [28,29]. The curve fitting analysis reveals the presence of symmetry in the P6/mmm space group, and the XRD data was analyzed using the Rietveld refinement approach [30]. Hence, the hexagonal structure was confirmed from the XRD and Rietveld refinement. Fig. 3 (a) represent the hexagonal crystal structure of WO₃. The figure displays the analysis of the XRD data, with the experimental data (I_obs) represented by black circles and the estimated intensities (I_cal) shown by red lines. The structural parameters acquired for the examined samples are displayed in Table 2. The disparity in ionic radii between the parent ion (W⁶⁺) and the dopant ion (Ag⁺) leads to a relatively small displacement of the XRD peaks towards higher 2θ values. As a result, as Ag doping concentrations increased, lattice parameters experienced slight contraction, except for 10% Ag: WO₃. The Debye-Scherer formula was employed to determine the average crystallite size of all samples by analyzing the reflected peaks in the X-ray diffraction patterns. It was assumed that the crystallite size is correlated with peak broadening. The estimated average crystallite size of all samples found was within the range of 25–28 nm. The size of crystallites and lattice strain have an impact on X-ray diffraction patterns. The W–H analysis is employed to differentiate between the deformation peak caused by the size and strain of the crystallite, considering the widening of the peak width as a function of 2θ. Scherrer method is most popular method for crystallite size estimation. However, this method is confined with limitation whether intrinsic strain is developed due to point defect, grain boundary triple junction and stacking faults [31,32]. Therefore, the Williamsons Hall method (W–H) and modified W–H method have been implemented here that includes the intrinsic strain along with the crystallite size [[33], [34], [35], [36], [37]]. It implies that the hexagonal structure of the WO₃ thin film is unaffected by Ag ions, and it cannot affect the crystalline phase of WO₃ [38].

Fig. 3.

Rietveld refinement plots illustrating the composition of pristine WO₃ and thin films containing 2, 4, 6, 8 and 10 at. % of Ag: WO₃.

Fig. 3 (a): Polyhedral hexagonal structure of WO₃ found from Rietveld refinement.

Table 2.

Calculated data from XRD analysis.

Sample	Lattice Parameters		c/a	V (Å3)	Average Crystallite Size, D (nm)	W–H Method UDM		${\mathrm{\chi }}^2$
	a = b (Å)	c (Å)				$D$ (nm)	$\epsilon$ (10−3)
Pristine WO3	7.32	3.81	0.52	176.80	27	46	1.8	4.1
2 at.% Ag:WO3	7.31	3.81	0.52	176.31	26	41	1.47	5.5
4 at.% Ag:WO3	7.30	3.81	0.52	175.83	28	72	2.93	6.3
6 at.% Ag:WO3	7.30	3.81	0.52	175.83	25	69	3.09	6.6
8 at.% Ag:WO3	7.30	3.81	0.52	176.31	28	71	2.68	5.8
10 at.% Ag:WO3	7.32	3.81	0.52	176.80	26	58	2.68	4.8

According to this model, the total broadening of XRD peak $\left({\beta }_{total}\right)$ is assumed as, ${\beta }_{total}={\beta }_{hkl}+{\beta }_{strain}$. In the following sections the structural analysis by both W–H and modified W–H method have been discussed and average crystallite size $\left(D\right)$ and microstrain $\left(\epsilon \right)$ have been estimated. The uniform deformation model (UDM) considers the strain that arises in nanostructures of a defective crystal due to its isotropic nature [35]. Here, physical broadening of the XRD profile due to intrinsic strain $\left({\beta }_{strain}\right)$ has been considered as, ${\beta }_{strain}=4\epsilon .\mathit{tan}\theta$ where $\theta$ is the Bragg position,

$$\mathrm{\beta }=\frac{0.9\lambda }{D}\times \frac{1}{\cos \left(\theta \right)}+4\epsilon .\mathit{tan}\theta$$ (6)

Now the further rearrangement of equation [6] is as,

$$\mathrm{\beta }*\cos \left(\theta \right)=\frac{0.9\lambda }{D}+4\epsilon .\mathit{sin}\theta$$ (7)

Now, the equation [7] has been represented graphically (Fig. 4) with the term $4\mathit{sin}\theta$ as independent variable (along X-axis) and $\beta *\cos \left(\theta \right)$ as a dependent variable (along Y-axis). This graph is a representation of the UDM model. The slope of linear fitting of this will provide the value of intrinsic strain and intercept gives average crystallite size $\left(D\right)$.

Fig. 4.

(UDM) The variation of βCos(θ) on the Y-axis and 4sin(θ) on the X-axis for the set of reflection data for different samples.

The lattice strain arises from the alteration of the lattice volume, which may either expand or shrink. This alteration is limited by the quantum size confinement that governs the behavior of the nanostructures. Consequently, the atomic configuration will undergo a modest modification in comparison to its original structure. Furthermore, the imposition of size restrictions will give rise to many imperfections within the lattice structure, thereby inducing lattice strain. The average crystallite size, as determined by the W–H model, lay within the range of [41–72] nm. This value is considerably larger than the size estimated by the Debye-Scherer formula, which incorporated the term of intrinsic strain to obtain the W–H model.

The original crystal, that is free of Ag has a certain value of intrinsic strain which reflects the effects of impurities and temperature. Introducing Ag to the crystal causes an increase in intrinsic strain values, these increases go proportionally for 4 and 6 at. % Ag dope WO₃. But it decreases slightly for 8 and 10 at. % Ag doped WO₃, which indicates a kind of relaxation inside the crystal. The bigger Ag⁺ (radius = 1.26 Å) ions absorb the W⁶⁺ (radius = 0.64 Å) ions inside the crystal when Ag concentrations rise. Hence, this process aims to attain the relaxation state by reducing the internal energy and total entropy of the crystal by relocating all the atoms throughout the whole gaps. The subsequent phase will include an increase in inherent strain due to an increase in the Ag concentration. Furthermore, this phenomenon also demonstrates the stochastic fluctuations in crystal size that occur when different concentrations of Ag are introduced, as the crystal attempts to optimize its energy and entropy by rearranging the locations of its atoms within the given space. Table 2 presents data on the intrinsic strain, lattice parameters, chi square, c/a ratio, and crystallite size ascertained using the Debye-Scherer formula and the W–H model.

The quantitative determination of the feature of the WO₃ and Ag: WO₃ thin film may be obtained using the texture coefficient (TC_hkl). This coefficient is estimated based on the X-ray diffraction (XRD) analysis of the (hkl) plane, using the following formula [17]:

$${TC}_{hkl}=\frac{\frac{I\left(hkl\right)}{I_0\left(hkl\right)}}{\frac{1}{N}\sum \frac{I\left(hkl\right)}{I_0\left(hkl\right)}}$$

The equation relates the measured intensity (I) of the WO₃ film to the standard intensity (I0) and the number of (hkl) diffraction peaks (N). An analysis of a randomly oriented crystallite shows a TC_hkl value of 1. If TC_hkl is more than 1, it indicates a higher prevalence of crystallite orientation. For varying concentrations of Ag, Fig. 3 shows the corresponding Texture Coefficient values for the three main diffraction planes (100), (001), and (200), and Table 3 lists these values. The TC value for the (100), (001), and (200) planes exhibit noticeable variations with increasing Ag-doping concentrations. These findings suggest a little re-orientation impact due to the Fe-doping ratio, which might explain the reduction in cracking thickness.

Table 3.

Texture coefficient of WO₃ and Ag: WO₃ samples.

Samples	Orientation Plane (hkl)	Texture coefficient TChkl
WO3	100	1.00
	001	1.16
	200	0.80
2% Ag: WO3	100	1.13
	001	0.97
	200	1.60
4% Ag: WO3	100	0.97
	001	1.04
	200	1.17
6% Ag: WO3	100	1.46
	001	0.99
	200	0.92
8% Ag: WO3	100	1.04
	001	1.08
	200	0.79
10% Ag: WO3	100	1.14
	001	1.09
	200	0.76

3.2. Surface analysis

FESEM images of pristine WO₃ and Ag: WO₃ thin films at 3k magnification and 5.0 kV are shown in Fig. 5. These coatings are deposited on substrates heated to 450 °C. The synthesis conditions and surface chemistry of WO₃ thin coatings can significantly affect their shape stability. The Ag: WO₃ FESEM micrograph shows granules with distinctive spherical shapes. Considering this, it is permissible for thin coatings of WO₃ to incorporate atoms of silver. Fig. 5. (a, b, c, and d) illustrate surface micro cracks. It has been observed that significant influence of thickness on fracture formation [39]. The widths of the cracks in diameter are shown in Table 4. However, no crack is observed for the higher percentage of Ag doping (especially 10% Ag doping). Merely cracks are found for 8% Ag doping. In Fig. 6 depicts the zoom view of crack formation of all samples except 10% doped Ag: WO₃ thin film.

Fig. 5.

(a–f) FESEM images of pristine WO₃ and 2, 4, 6, 8 and 10 at. % Ag: WO₃ thin films at $3\mathrm{k}$ magnifications.

Table 4.

Average particle size and crack width for each as-deposited thin film.

Sample	Average particle size in diameter, (μm)	Crack width, (μm)	Standard deviation
Pristine WO3	3.40	0.33	0.10
2% Ag	3.41	0.11	0.04
4% Ag	3.62	0.10	0.02
6% Ag	3.55	0.15	0.06
8% Ag	3.79	0.07	0.01
10% Ag	1.85	–	–

Fig. 6.

(a–e) Zoom view of FESEM images for pristine WO₃ and 2 to 8 at. % Ag: WO₃ thin films.

The addition of Ag produces sphere grains and reduces the crack's thickness. These fractures result from an imbalance between the compressive stress, thermal expansion of WO₃ thin coatings and the glass substrate. Introducing Ag by doping can incorporate impurity atoms into the crystal lattice of the film material. Impurity atoms can immobilize dislocations, which are imperfections in the arrangement of atoms in a crystal structure that can serve as starting points for cracks. Ag doping can significantly decrease the mobility of dislocations by immobilizing them, hence reducing the propagation and growth of fractures. According to our findings, when the concentration of Ag doping is increased, cracks would become invisible.

Ag dopant is incorporated into WO₃ thin films as micro-particles of Ag and W, combined on the surface of Ag: WO₃ thin films. The particle size increased as Ag concentration increased. Image of 10 at.% Ag: WO₃ film reveals the formation of incompatible structure grains and spheres. It is likely because a higher concentration of Ag could disrupt the crystal matrix and impede the thin film's growth.

The higher the concentration of Ag (10 at.%), the grains become the smaller due to Ag's collision with W. In addition, increasing particle size leads to increased absorbance [40]. As the particle size decreases, the absorbance of 10 at.% Ag-doped material has shown the decline nature. For particles in a fluid, such as powder or granular material, the GSD is an ensemble of numbers or mathematical functions that characterizes the distribution of particle sizes. Fig. 7 depicts the histogram of the GSD for all synthesized samples. From FESEM pictures, ImageJ software was used to reveal a histogram for GSD.

Fig. 7.

Histogram of the Gaussian size distribution (GSD) of pristine WO₃ and Ag: WO₃ thin films.

3.3. Elemental analysis

For the as-deposited film, the EDX method was used for the quantitative examination of the films. Fig. 8 shows the results of an EDX analysis performed on all samples, confirming the existence of W and O.

Fig. 8.

EDX spectrum of thin films made of (a) pristine WO₃, (b) 2, (c) 4, (d) 6, (e) 8, and (f) 10 at.% Ag: WO₃.

These elements can be seen as two distinct peaks, with respective mass percentages of 74.1 and 25.9. Moreover, the spectrum reveals no impurity originating from carbon additives. Ghasemi et al. [41] reported an equivalent EDX analysis outcome. In thin films of pristine WO₃, the proportion of W is significantly greater than that of O. The EDX Fig. 8 for doping of Ag confirmed the presence of Ag at the 3.00 eV position. The mass percentages (mass%) of W, O and Ag are precisely understandable through the graphical representation in Fig. 9.

Fig. 9.

Mass percentages of every element of deposited sample for different Ag concentration.

3.4. Optical properties

Data from a UV–Vis absorption spectrometer has been used to investigate the samples' optical characteristics. This absorbance graph showed that the samples have slight variations from one another. Absorbance changes significantly when Ag dopants are introduced.

3.4.1. Absorption coefficient analysis

The optical absorption coefficient of all samples are shown in Fig. 10 using UV–Vis spectra. In Fig. 10 without 10 at.% Ag doping, the absorption coefficient of the samples rises with doping concentrations of Ag. Surprisingly, the absorption coefficient of WO₃ thin films doped with 10 at.% Ag is lower than films doped with 6 and 8 at.% Ag. When measured in wavelength units, absorption coefficient begins at 360 nm. As the wavelength rises, both films become less absorption coefficient. This drop in absorption coefficient is characteristic of a material having an optical band gap [42]. Therefore, Ag: WO₃ thin films have higher absorption coefficient in the visible region than their pristine counterparts. WO₃ thin films with higher sensitivities to visible light due to Ag contamination may be susceptible to photo-degradation, according to this evidence. Since absorption coefficient declines with increasing silver incorporation, the transparency of the films gradually decreases. There are several potential causes for the decrease in transmittance that accompanies the increased incorporation of silver. The presence of Ag nanoparticles may cause a reduction in transmittance due to Surface Plasmon Resonance (SPR). A phenomenon known as SPR occurs when the collective oscillation of CB electrons occurs in resonance with the oscillating electric field of incoming light, causing the non-radiative excitation of strong plasmonic electrons. Ag inclusion causes this behavior. A few other materials, like gold and aluminum, are also responsible for this phenomenon. This phenomenon has been reported by several publications as a consequence of Ag inclusion in WO₃ thin films [43,44].

Fig. 10.

Absorption coefficient of pristine and Ag: WO₃ thin films for different wavelengths.

3.4.2. Optical band gap

The energy difference between the bands has been calculated using UV–Vis Spectrometry. Optical absorbance has been used with the Tauc equation to determine the optical energy band gap of the films.

$$\alpha h\nu =A{\left(h\nu -E_g\right)}^n$$ (8)

Here, α represent the absorption coefficient, ℎν depicts the photon energy, n is the parameter connected to the distribution of the density of states, and A is a constant or Tauc parameter [15].

The film thickness (t) is determined by measuring the displacement (h) of the fringe system over the film substrate step and using the relation [45],

$$t=\frac{h}{fringsspacing}\times \frac{\lambda }{2}$$

where λ is the used monochromatic light's wavelength. We may write as follows if the fringe spacing is (l),

$$t=\frac{h\lambda }{2l}$$

Fig. 11 indicates the linear band gap plots for (αℎν)² vs E (eV). Here, the direct nature n is allowed to be 1/2, and the indirect nature n is allowed to be 2. An electron that is located in the CB of a direct nature band gap semiconductor has the ability to travel to an empty state in the VB, thereby releasing the energy difference. In comparison, an electron in the lowest part of the CB of an indirect band gap semiconductor cannot go straight down into the highest part of the VB. But along with its energy, it must also change its momentum. The linearity of these plots indicates the transitions in these films are indirect. Using the intersection of the plot's tangent and the plot, the band gap value has been determined. The literature and the findings we reported are consistent regarding the indirect band gap of 3.2 eV for pristine WO₃ thin films [12,43]. In the range of 0–8 at.% Ag doping concentration, the band gap of a thin WO₃ film decreases from 3.2 to 2.9 eV, respectively. The calculated value of band gap and crystallite size of pristine and WO₃ with varying amounts of Ag dopant are shown graphically in Fig. 12. Table 5 represents the calculated band gap and thickness of the all samples.

Fig. 11.

Tauc's plot of (αℎν)² vs E (eV) of pristine and doped WO₃ thin films with different Ag concentrations.

Fig. 12.

Graphical representation of optical band gap and crystallite size for various Ag concentrations on WO₃ thin films.

Table 5.

Bandgap and thickness of all synthesized thin films.

Sample	Bandgap, (eV)	Thickness of the sample, (nm)
Pristine WO3	3.20	635 $\pm$ 20
2% Ag	3.15	631 $\pm$ 20
4% Ag	3.07	638 $\pm$ 20
6% Ag	2.95	627 $\pm$ 20
8% Ag	2.90	636 $\pm$ 20
10% Ag	3.02	641 $\pm$ 20

Narrowing the band gap and visible light absorption could be happened due to the oxygen vacancies, and the following interpretation of these data is also possible: Ag-doping creates new localized energy states between tungsten's valence and conduction bands at low at% of doping (up to 8%). These localized energy states function as newly introduced trapping levels, which cause Ag-doped tungsten trioxide thin films' bandgap to decrease. For 10 at.% Ag: WO₃ thin film has a band gap value of 3.02 eV. This change, a red shift in Fig. 10 aligns with absorption data. The results have shown that adding Ag to WO₃ thin films could have an essential effect on reducing the impact of the band gap energy. Several past studies agree with this observation of decreasing band gap energy with indirect bandgap [20,46,47].

3.5. Redox potential

The charge carrier generation and redox potential are crucial for understanding the semiconductor materials' water splitting (Photo-catalytic Activity). The position of CBM (Conduction Band Maximum) VBM (Valance Band Minimum) of the synthesized pristine and Ag: WO₃ thin films are determined for studying the redox potential for this present study.

The semiconductor band-edge locations are key factors for identifying the energetics of PEC (Photo Electrode Chemical) water-splitting devices [48]. The following equations are utilized to compute the band edge of VB and CB of our synthesized semiconductor catalysts [49].:

$$E_{CB}=X-E_e-\frac{1}{2}{E}_g$$ (9)

$$E_{VB}=E_{CB}+E_g$$ (10)

Here, $E_g$, the calculated energy of band gap is calculated from the plot (using equation (10)), where $E_{CB}$ and $E_{VB}$ are the band edge potentials of CB and VB, respectively, X represents the calculated Mulliken electro-negativity, $E_e$ is the energy of the free electron on the hydrogen balance (4.5 eV), and so on. In light of the above equation, Fig. 13 illustrates the computed $E_{CB}$/ $E_{VB}$ values of pristine and Ag: WO₃ films and the potential band gap concerning the NHE (Normal Hydrogen Electrode) (pH = 0). Given that the VB edge potential (O₂/H₂O) is 1.23 eV and the CB edge potential (H⁺/H₂) is 0 eV, the theoretically lowest band gap for water electrolysis is 1.23 eV. Different Ag doping concentrations on WO₃ thin films cause the edge of the CB and valence band to change. The downward movement of the CB's leading edge brings it closer to the H⁺/H₂ reduction potential.

Fig: 13.

Schematic diagram of positions of CBM and VCM of pristine and different concentration of Ag: WO₃ thin films concerning the normal hydrogen electrode (NHE).

In Ag₂O/WO₃ crystals, incident photon energies (ℎν) are absorbed by the n-type WO₃ semiconductor, which then excites electrons from the WO₃ VB to the WO₃ CB and causes them to migrate to the p-type Ag₂O semiconductor, facilitating electron-hole separation. For a semiconductor photo-catalyst, edge of valence band (VB) must have positive than oxidation potential of O₂/H₂O. In contrast, the CB edge position must have less than zero [50]. In this study, Ag dopant has been effective for tuning the band edge potential and enhancing the photo degradation process of the WO₃ thin films.

When the semiconductor photo-catalyst is exposed to radiation ($h\mathbf{ʋ}$), the hole ${(h}^{+}$) and electron $\left(e^{-}\right)$ are created for jumping the electron VB to CB. Those hole and electron are responsible for forming free radicals, and the free radicals (hydroxyl) are oxidative. For producing free radicals spontaneous photo-degradation process is occurred.

3.6. Electrical properties

3.6.1. I–V curve

I–V characteristic curve also known as current-voltage curves describing how an electrical device or component operates in an electrical circuit. The voltages used to measure the current and voltage drops are examined in the range of 10–65 V. For various voltages, the current passing through pristine WO₃ thin films is less than that passing through Ag: WO₃ thin films. Hence, Ag is a highly conductive material, adding Ag to WO₃ thin films causes the current flow to rise.

I–V curves are practically linear and show the ohmic nature, as depicted in Fig. 14. The results indicate that the rising doping concentrations of Ag ions enhance the current flow. This ohmic behavior of WO₃ thin films is comparable to prior studies [51].

Fig. 14.

I–V Characteristic curves for pristine WO₃ and Ag: WO₃ thin films.

3.6.2. Variation of resistivity with temperature

Ag doping substantially reduces the resistivity of pristine WO₃ thin films at room temperature, which is $19.63\times {10}^4$ Ω-cm. Resistivity results from a trade-off between two simultaneously completing processes [52]. It is believed that the hopping conduction process controls the electronic transport of WO₃, and electrons are the primary carriers of oxygen vacancies. Fig. 15 has depicted the temperature dependence electrical resistivity of each sample. Temperature-dependent decreases in resistivity reveal the behavior of semiconductors is the Negative Temperature Coefficient (NTC). The semiconductors with NCT utilize it to understand the signal for temperature changes.

Fig. 15.

Temperature vs resistivity curves of pristine and Ag: WO₃ thin films.

3.6.3. Activation energy

The four-point probe method was used to characterize the temperature dependence conductivity of pristine and Ag: WO₃. This work has shown that pristine and Ag-doped samples affect electrical properties significantly. The following Arrhenius law was applied,

$$\mathrm{\sigma }={\sigma }_0\exp \left(\frac{-E_{act}}{kT}\right)$$ (11)

In the following equation, $\mathrm{\sigma }$ represents conductivity, ${\sigma }_0$ represents the pre-exponential factor, E_act represents the activation energy for conduction, k represents the Boltzmann constant, and T represents the absolute temperature.

Equation [13] can be expressed as,

$$In\sigma =\left(\frac{-\Delta E}{K_{B}T}\right)+In{\sigma }_0$$ (12)

The slopes of $\ln \sigma$ vs $\frac{1000}{T}$ have plotted using the Arrhenius relation to determine the activation energy. A plot of $\ln \sigma$ vs $\frac{1000}{T}$ is shown in Fig. 16. From the resistivity curve, it has been confirmed that higher temperatures result in higher conductivity; similarly, the activation energy graph as well [53]. The calculated value of activation energy is enlisted in Table 6.

Fig. 16.

The changes in activation energy for pristine and Ag: WO₃ thin films as the temperature varied.

Table 6.

Calculated value of activation energy.

Sample	Activation energy, $\Delta$ E1 (eV) (308K–348K)	Activation energy, $\Delta$ E2 (eV) (353K–393K)
Pristine WO3	0.74	0.89
2% Ag:WO3	0.68	0.92
4% Ag:WO3	0.80	0.89
6% Ag:WO3	0.67	0.83
8% Ag:WO3	0.58	0.78
10% Ag:WO3	0.64	0.71

Thus, the activation energy $\Delta E$ can be represented by the following equation,

$$\Delta E=-\frac{\ln \sigma }{\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$T$}\right.}{K}_B$$ (13)

When reactant molecules are heated, their mobility increases, leading to more frequent and forceful collisions taking place within individual molecules between the atoms and bonds. This increases the likelihood of bond rupture, which is necessary for the chemical reaction to occur. For bonds to be broken, the molecule must enter the unstable transition state. This state requires activation energy to be added to the molecule [54]. However, because the transition state is unstable, reactant molecules quickly move onto the next stage of the reaction rather than remaining in the transition state for long periods.

4. Conclusions

A simple spray pyrolysis method was used to deposit WO₃ thin films onto glass substrates and revealed both pristine and Ag: WO₃ films are to be polycrystalline with a hexagonal crystal structure. The addition of Ag atoms leads to an inconsistent calculation of crystallite size using both the Debye-Scherer formula and the W–H plot. The crystal lattice defect is responsible for the maximum lattice strain seen in WO₃ doped with 6 at. % of Ag. Thin films of pristine WO₃ grew in the (200) direction while doping with Ag caused a small shift in the select orientation plane. Different amounts of Ag doping were found to significantly alter the surface morphology of WO₃, as observed by FESEM. The absorption spectrum was consistently affected by the particle size distribution. UV–Vis spectroscopy revealed a red shift and a reduction in the width of the optical band gap, which is particularly significant for optoelectronic applications. Furthermore, the adjustment of conduction and valence band edges by Ag doping has a substantial impact on the photocatalytic process and 2 at. % Ag-doped WO₃ exhibits notable performance in the photocatalytic process. Both pristine and Ag: WO₃ thin films exhibit a negative temperature coefficient behaviour, making them suitable for use as heat sensors. Based on a thorough review of this work, it is evident that Ag has the potential to be an appropriate candidate for doping WO₃ thin films to enhance their performance in optoelectronic devices.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data statement

The data used for this article is derived from the experiments. These are not available online.

CRediT authorship contribution statement

Md Arifuzzaman: Writing – original draft, Validation, Investigation, Formal analysis, Data curation. Tusar Saha: Validation, Methodology, Investigation, Formal analysis, Data curation. Jiban Podder: Writing – review & editing, Validation, Supervision, Methodology, Investigation, Data curation, Conceptualization. Fahad Al-Bin: Validation, Formal analysis, Data curation. Hari Narayan Das: Validation, Formal analysis, Data curation.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.